Shabnam Kazembalaghi et al., Digital equity and government support during COVID-19 (2024)
Summary
A research article by Shabnam Kazembalaghi, Jerry Coakley and José M. Liñares-Zegarra of Essex Business School and Silvio Vismara of the University of Bergamo, Digital equity and government support during COVID-19, in Small Business Economics (2025) 64:1679-1705, DOI 10.1007/s11187-024-00961-9, accepted 8 August 2024 and published online 5 September 2024. The authors find that UK equity crowdfunding grew during the pandemic, with funding amounts up 15% and investor numbers up 32% on Crowdcube, as digital due diligence replaced in-person meetings with business angels and venture capital funds. They argue that government-backed loan guarantee schemes, including those with full repayment guarantees, acted as a quality signal to investors and a financial lifeline for seed firms, so that equity became the funding of first choice rather than last resort.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Cite this document
“Shabnam Kazembalaghi et al., Digital equity and government support during COVID-19 (2024).”
Original at link.springer.com. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/shabnam-kazembalaghi-et-al-digital-equity-and-government-support-durin-e2b18be754/
, SHA-256 5a9c3f48c6a3c6b1e7723ddf737d4daad905b9d8a2cf2622c57d10e08f815c33.Full text
Small Bus Econ (2025) 64:1679–1705
https://doi.org/10.1007/s11187-024-00961-9
RESEARCH ARTICLE
Digital equity and government support during COVID‑19
Shabnam Kazembalaghi · Jerry Coakley ·
José M. Liñares‑Zegarra · Silvio Vismara
Accepted: 8 August 2024 / Published online: 5 September 2024
© The Author(s) 2024
Abstract The advent of COVID-19 portended a small firms during a period of heightened economic
dire liquidity crunch for small firms as traditional uncertainty.
external funding sources were severely curtailed.
Defying expectations, initial equity crowdfunding Plain English Summary During the pandemic,
(ECF) campaigns not only withstood the pandem- when physical meetings were restricted, UK equity
ic’s onslaught but also saw unprecedented growth in crowdfunding (ECF) platforms became a crucial
funding volume, investor participation and overfund- lifeline for start-ups, enabling equity financing with-
ing. The upshot was that external equity, the tradi- out face-to-face interactions with investors like busi-
tional funding of last resort, became the first choice. ness angels. ECF offerings flourished with funding
Increased ECF funding especially for seed ventures amounts increasing by 15% and investor numbers by
are likely linked to government-backed loan guaran- 32%. Their success was due to digital due diligence
tee schemes that acted as a quality signal for inves- mechanisms that effectively identified high-quality
tors. The paper highlights the unanticipated positive ventures. Moreover, government loan guarantee
synergies between public support mechanisms and schemes, including those with full repayment guaran-
private equity dynamics where equity was funding tees, provided significant support. Initially, there were
of first choice for many small firms seeking exter- concerns about these schemes negatively affecting
nal funding. These developments underscore ECF’s other finance sources. However, they proved to be a
central role in digitally channelling equity capital to financial lifeline for small firms, enhancing their via-
bility and improving their chances of raising equity
capital through Crowdcube. As a result, innovative
S. Kazembalaghi · J. Coakley · J. M. Liñares‑Zegarra seed firms thrived during COVID-19 as government
Essex Business School, University of Essex, Colchester, aid in the pandemic’s first year guaranteed their sur-
UK vival. The main implication is that ECF – a very risky
e-mail: sk21214@essex.ac.uk source of finance – flourished during COVID-19
J. Coakley through digital flexibility and government support.
e-mail: jcoakley@essex.ac.uk
J. M. Liñares‑Zegarra Keywords Equity crowdfunding · COVID-19 ·
e-mail: jmlina@essex.ac.uk Digital finance · Seed firms
S. Vismara (*)
University of Bergamo, Bergamo, Italy JEL Classification E44 · G01 · G18 · G38 · L26
e-mail: silvio.vismara@unibg.it
Vol.: (0123456789)
1680 S. Kazembalaghi et al.
1 Introduction and venture capital (VC) funds and partly because a
digital platform became an attractive outlet for pro-
The COVID-19 pandemic was a major exogenous fessional investors whose traditional day-to-day
shock to all economies worldwide and it raised the modus operandi involving in-person meetings was
spectre of a doomsday scenario for small firms hit by curtailed during the pandemic. The COVID-19 pan-
a combination of lockdown restrictions and an obvi- demic and its associated lockdown restrictions had a
ous lack of external funding sources (Belitski et al., negative impact on both BA and VC initial funding
2022). The severity of the economic downturn and the rounds (Mason & Botelho, 2021). This adverse effect
scale of the related mortality numbers were unprec- may have been exacerbated by the limitations on in-
edented in recent decades (Altig et al., 2020; Baker person meetings used by professional investors for
et al., 2020). In the UK, the COVID-19 pandemic and due diligence, as previously noted by Harrison et al.
related government lockdown restrictions impacted (2016) in the pre-pandemic context. All this took
both new startup formation and the survival prospects place against a bleak mainstream financial market
of existing ventures (Altig et al., 2020; Brown & backdrop, where the FTSE 100 and the pound sterling
Cowling, 2021; Brown et al., 2020; Calabrese et al., posted record-breaking losses (Paterson et al., 2024).
2022). This paper examines how equity crowdfund- The digitalization process was the key to ECF success
ing (ECF) as a leading form of digital finance (Block during COVID-19 as argued below.
et al., 2018, 2021) helped young small firms meet A key feature of the recent (post-2016) UK digital
the complex challenges the pandemic posed using equity model is that the ECF campaign lead investor
a sample of 660 successful Crowdcube initial ECF undertakes her own due diligence before making a
campaigns, 2018–2023. The findings reveal an unex- sizeable investment. She also has to organise a syn-
pected surge in ECF performance during COVID-19. dicate to commit 20% of the target—a new provi-
The amount of equity raised, number of funders and sion point mechanism (PPM)—before the campaign
the degree of overfunding (amount/target) all rose by can go public (Coakley et al., 2024). Here the impact
double-digit figures in contrast to underperforming of the lead investor—typically a BA or VC—is akin
public equity activity (Paterson et al., 2024) and to to that of a third-party prior financing effect (Klein-
the pecking order theory that views external equity as ert et al., 2020). These factors imply that low-quality
funding of last resort. ventures typically fail to meet the PPM while success-
This paper builds on recent calls to explore the ful high-quality ventures attract professional inves-
impact of digitalization on resource mobilization tors even in times of increased uncertainty (Fisch &
by focusing on ECF as a form of digital equity (e.g. Momtaz, 2020). Conveniently, lockdown measures
Estrin et al., 2018; Inceoglu et al., 2024) against a also prompted BA and VC funds to shift their invest-
background of an upward trend in professional equity ment towards ECF platforms (Mason & Botelho,
investment from 2011. Kacer and Wilson (2023) 2021). Concurrently, VC involvement in ECF cam-
provide a great overview of equity investment value paigns had increased as a result of their diversifica-
and number of deals in the UK over the 2011–2022 tion (“spray and pray”) strategy over the past decade
period. They find that both measures of equity invest- where they invested smaller amounts of funding and
ment trended upward since 2011 to reach a peak in governance across a broader array of startups (Ewens
2020Q1 (beginning of COVID-19). However, they et al., 2018).
declined in 2020Q2 before recovering in 2021. Deal Digital ECF platforms are an ideal outlet in this
value reached a new peak in 2022Q1 before it then context because their ownership and exit rights are
declined. almost identical to those of typical VC syndicates
The paper makes two contributions to the litera- but within a digital framework. The digitalization of
ture. The first is that it explains why ECF, as a form equity finance played a pivotal role in the boost to
of digital equity, generally flourished during COVID- ECF performance during COVID-19 with platforms
19 as part of the general upward trend in equity predominantly utilizing pre-campaign big data ana-
investment in the UK. This was partly because digital lytics for due diligence, facilitating online funding
platforms proved more adaptable and resilient than through ECF platforms and implementing post-cam-
traditional equity providers like business angels (BA) paign digital ownership, monitoring and corporate
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1681
governance systems (Coakley et al., 2024; Inceo- Interruption Loan Scheme (CBILS).2 The 80% gov-
glu et al., 2024; Kleinert et al., 2022). The resultant ernment guarantee and lender-set interest rates on
increased involvement of professional investors on CBILS still enabled lenders to carry out their stand-
ECF platforms triggered cascading effects among ard due diligence and vetting of borrowers. Borrow-
the crowd and contributed to enhanced ECF perfor- ers had to provide documents to show they could
mance (Vismara, 2018). The increase in ECF funding afford to repay the CBILS loan, such as management
of small firms also reflects a broader trend in West- accounts, cash flow forecasts, business plans and
ern economies involving a shift from public to private details of assets (BEIS and BBA, 2020). The lend-
equity and where innovative young firms benefit from ers remained responsible for their lending decisions
staying private longer (Ewens & Farre-Mensa, 2022; under CBILS. Moreover, the lenders included not
Vismara et al., 2012). only the traditional High Street banks but also Fin-
The paper’s second contribution is that it high- tech lenders such as marketplace (P2P) platforms like
lights the role of enlightened public intervention in Funding Circle and neobanks like Starling. Fintech
the form of a battery of government policy initiatives lenders accounted for a large proportion of LGS loans
in boosting the performance of ECF firms—and seed largely due to their digital due diligence systems that
stage firms in particular—in the critical first year of resulted in fast decisions to firms.
COVID-19. In particular, loan guarantee schemes The increased demand for equity capital by small
(LGS) provided a funding framework to stabilize the firms is consistent with equity as funding of first
financing of small and, especially, seed firms at the choice for the smallest firms and with the theoreti-
height of the pandemic. Government-backed loans cal model predictions of Fulghieri et al. (2020). The
played a critical role here as they provided a quality ECF literature provides some evidence of ECF as
signal that small firms could meet their financial obli- funding of last resort as the POT predicts (Blaseg
gations when they become due. On the other, ECF et al., 2021; Walthoff-Borm et al., 2018) but their
platforms were ceteris paribus more likely to view samples do not extend beyond 2015. In other words,
LGS-supported firms as being less risky (due the gov- they applied to the period of pure ECF (Vismara,
ernment guarantee and first-year interest (fees) con- 2016) with mainly crowd investors when serious
cession) than equivalent firms without an LGS loan. adverse selection problems persisted because due
Seed firms are considered distinctive due to their diligence was hampered by the collective action
higher innovation levels, presence in high-growth problem. Finally, the Fulghieri et al. (2020) model
sectors and a stronger emphasis on cutting-edge tech- predicts that firms with pre-existing debt are likely
nology compared to other small firms. Kleinert et al. to use equity in follow-on offerings. An unin-
(2020) stress that prior financing by BA or VC exerts tended consequence of government-backed loans at
a strong signaling effect on seed firms. The latter also a time of falling equity valuations was an increase
have lottery stock properties (Bali et al., 2011) such in capital gearing ratios. The recent Cowling and
as right-skewed returns which make them highly Wilson (2023) findings provide valuable insights
attractive to professional investors like VCs. into how rising inflation can influence firms’ deci-
In this context, the government LGS helped young sions to repay guaranteed loans ahead of sched-
startups and seed firms in particular to overcome the ule. By paying back debt early, firms may enhance
liquidity crisis and related challenges arising during
the pandemic until March 20211 (Brown & Cowl-
ing, 2021). The LGS included the Bounce Back 2
During COVID-19, the UK government introduced three
Loan Scheme (BBLS) and the Coronavirus Business major Loan Guarantee Schemes (LGSs) two of which aimed at
supporting small firms, namely the Bounce Back Loan Scheme
(BBLS) and the Coronavirus Business Interruption Loan
Scheme (CBILS). These three initiatives helped over 1 mil-
lion UK businesses, with BBLS being the most widely utilized
1
LGS schemes in the UK were extended until June 2021 but (Cowling et al., 2023; Cowling et al., 2022). For further details
only for asset finance lending as global supply chains delayed on BBLS and CBILS, see: https://www.gov.uk/guidance/apply-
the delivery of the asset. We are grateful to an anonymous ref- for-a-coronavirus-bounce-back-loan and https://www.gov.uk/
eree for this point. In total they contributed some £1.608 m to guidance/apply-for-the-coronavirus-business-interruption-loan-
small firms. scheme
Vol.: (0123456789)
1682 S. Kazembalaghi et al.
their attractiveness to equity investors and improve an overview of the methodology and describes the
access to equity financing options in the future.3 empirical models employed to test the hypotheses.
Successful ECF campaigns can then directly assist Section 4 analyses and discusses the empirical find-
the firms in recapitalizing and moving their gear- ings. The final section concludes.
ing ratios back towards tolerable levels. The role of
ECF funding in this context sharply contrasts with
the pre-COVID-19 funding of seed firms. Histori- 2 Literature review and hypotheses
cally, BA and VC funds would invest in seed firms
via convertible (loan) notes4 that offer no immediate 2.1 Funding responses to natural disasters
ownership or control rights but could be converted
to equity at a future Series A round (Weiss, 2023). The crises following natural disasters typically have
The findings on the outperformance of ECF cam- the most detrimental impact on smaller and younger
paigns during the COVID-19 period link to the firms’ day-to-day business and on their ability to
related literature. They support the early COVID-19 access traditional funding sources such as bank lend-
period ECF findings of Vu and Christian (2023) for ing (Cortés & Strahan, 2017; Nguyen & Wilson,
the UK and the Cumming et al. (2021b) ECF and 2020). Like a natural disaster, COVID-19 represents a
P2P lending results for the USA. It validates the con- major exogenous shock (Baltas et al., 2022; Chandler
jecture of Walthoff-Borm et al. (2018) that ECF can et al., 2021; Li et al., 2023) and this setting is used to
theoretically serve as the primary funding source for explore shifts in the digital capital raising outcomes
private firms during crises (Guenther et al., 2018) as of smaller firms. By contrast to a natural disaster, a
it fosters founder control and provides value-added pandemic involves the outbreak of an infectious dis-
services like digital corporate governance to investors ease over entire economies—the world in extremis—
(Estrin et al., 2018). that generally affects a significant proportion of the
The paper’s seed firm findings link to the related population and leaves huge numbers of fatalities in its
COVID-19 literature. Calabrese et al. (2022) exam- wake over a prolonged period. The related economic
ined the role of COVID-19 government LGS in crises lead to more extreme information asymmetries
supporting UK SMEs and established that the most impacting particularly young startups, making access
significant proportion of loans were directed to bet- to external—including digital—finance even more
ter-performing micro and small businesses. Brown challenging. However, recent research indicates that
et al. (2020) found that seed firms accounted for most alternative financing might exhibit distinct behav-
of the equity funding of small firms during COVID- ioural patterns during periods of heightened uncer-
19. Our paper also contributes to the literature on tainty (Baltas et al., 2022; Zhang et al., 2015).
the effectiveness of UK government LGS during the COVID-19 highlighted a sharp contrast between
COVID-19 crisis (Cowling et al., 2022, 2023; Wilson the modus operandi of ECF and traditional sources
et al., 2023). ECF outperformance was particularly of equity for small firms. Due diligence by traditional
notable during the government LGS period, suggest- VC and BA investors relies on face-to-face contact
ing a direct and positive impact of government inter- over a period of time to perform detailed evaluations
vention on crowdfunding success, particularly among of prospective startups seeking equity (Cowling et al.,
seed firms. 2021; Cumming et al., 2021b). By contrast, ECF
The remainder of this paper is organised as fol- platforms rely heavily on digital and and also word
lows. Section 2 reviews the literature and outlines of mouth (for example, networks of accountants and
the hypotheses to be examined. Section 3 provides lawyers that deal with such ventures) due diligence.
Consequently, many professional investors migrated
their investments to online ECF platforms during the
3
We are grateful to an anonymous referee for this insight. pandemic (Mason & Botelho, 2021). This was a gen-
4
Detailed information on VC investments is extremely dif- eral trend that had earlier gained momentum with BA
ficult to obtain. See this link for UK VC. https://www.osbor (Wright et al., 2015) and with the “spray and pray”
neclarke.com/insights/vc-focus-why-vcs-and-companies-use-
convertible-loan-notes-and-bridging-rounds-uk. Weiss (2023) venture capital strategy over the past decade where
confirms a similar pattern in the USA. small investments are spread over a wider range of
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1683
startups (Ewens et al., 2018). Note that such invest- et al. (2018) and Blaseg et al. (2021) find that ECF
ments in an ECF campaign by BA or VC would serve was the funding of last resort for ECF samples of UK
as a strong prior financing signalling effect leading and German firms, respectively, which is in line with
to potential cascading behaviour by other investors the traditional POT. However, their ECF samples
(Meoli & Vismara, 2021; Vismara, 2018). end in 2015 and so belong to the pure ECF era (Vis-
mara, 2016) where small investors dominated ECF
2.2 Pecking order theory campaigns giving rise to the collective action prob-
lem where investors tend to free ride on others. By
The traditional pecking order theory (POT) pos- contrast, the post-2016 syndicated ECF model with a
its that firms prefer internal to external funds due to lead investor (Coakley et al., 2024) has transformed
adverse selection costs (Myers & Majluf, 1984) and, the role of professional investors in ECF as it has lev-
among external funds, they prefer debt to equity. POT eraged their due diligence and monitoring skills to
was developed at a time when public equity—given attract high-quality investors to their campaigns. In
by stock exchange flotation—was in the ascendancy. particular, the provision point mechanism of requir-
However, recent decades have witnessed a move in ing the lead investor syndicate to commit to a mini-
the opposite direction with a fall in public (listed) mum of 20% of the target prior to the campaign going
firms and a simultaneous rise in private firms (Ritter public is a game changer in terms of sharply improv-
et al., 2013). The empirical study of Lattanzio et al. ing the quality of small firms permitted to run initial
(2023) confirms this trend for developed economies ECF campaigns. Moreover, Kleinert et al. (2022)
and shows that it emerged earlier (in the mid-1990s) provide survey evidence that the stringent due dili-
in the UK and USA relative to other developed econ- gence standards imposed by ECF platforms across a
omies. The shift in Western economies from pub- range of countries imply that only high-quality ven-
lic to private equity is driven by factors such as the tures are permitted to run ECF campaigns. For such
increased costs of regulation on stock markets (public firms, ECF is funding of first resort. In this context,
equity) and the fact that the deregulation (e.g. allow- Walthoff-Borm et al. (2018) were prescient in arguing
ing the establishment of ECF platforms) of the pri- that ECF could theoretically be funding of first resort
vate equity markets enhances the benefits to small as it permits founder control and offers value-added
firms of remaining private (Ewens & Farre-Mensa, services like visibility and feedback.
2022; Momtaz, 2023).
The COVID-19 pandemic had a dual effect on the 2.3 Hypothesis
equity financing of firms. On the one hand, it exerted
a direct effect on the professional investors (VC and The development and growth of digital finance plat-
BA) that traditionally supplied private equity but forms such as ECF, marketplace lending and other
who suffered due to lockdown restrictions (Brown digital finance providers (e.g. market invoice or the
et al., 2020; Mason & Botelho, 2021). More gener- prepayment of invoices for a fee) began in the wake of
ally, COVID-19 led to sharply increased uncertainty the Great Financial Crash. The COVID-19 lockdown
for public equity (Brown & Rocha, 2020) which restrictions curtailed traditional investor activities
drove down stock exchange valuations and ham- which relied heavily on face-to-face due diligence.
pered the IPO markets. The latter are quite sensitive This provided ECF platforms with a unique oppor-
to the negative mood in the markets. Paterson et al. tunity to attract more professional investors, thereby
(2024) report that the FTSE 100 and pound sterling enhancing campaign performance through a BA or
witnessed record-breaking losses during the COVID- VC certification effect. These investors responded by
19 period that would have depressed the IPO market investing in ECF either as lead investors in syndicated
also. In other words, public equity finance was strictly ECF campaigns or by diversifying their investment
rationed during COVID-19. risk across a range of ECF campaigns in a period of
On the other hand, the pandemic-induced shift heightened risk. Additionally, they were drawn to the
from traditional to digital finance is clearly illustrated bespoke value-added services provided by the plat-
by the performance of ECF campaigns. How does form (Walthoff-Borm et al., 2018). These include a
this link to POT? Researchers such as Walthoff-Borm digital nominee corporate governance structure that
Vol.: (0123456789)
1684 S. Kazembalaghi et al.
monitors ECF firms and protects the ownership and the COVID-19 pandemic differed from those of pre-
exit rights of all investors (Coakley et al., 2024). vious crises. Notably, there was an initial decline in
The COVID-19 boost to digital finance is exem- insolvency rates, particularly in the first year of
plified by initial ECF campaigns run by high-qual- COVID-19. In such circumstances, it is highly likely
ity startups during the pandemic period. These new that, ceteris paribus, ECF platforms would target
equity injections would help to boost their funding small firms with a government-back loan over those
performance relative to those in the pre-COVID-19 without such backing. This leads to the following
period as outlined in H1A: hypothesis on the government LGS and their impact
on ECF campaigns.
H1A: Initial ECF campaigns launched during
the COVID-19 period outperform those of the pre- H2A: The availability of government loan guaran-
COVID-19 period. tee schemes in the first year of COVID-19 positively
impacted ECF campaign performance.
During the post-COVID-19 period, the crisis fac-
tors that previously boosted ECF outcomes no longer The LGS targeted small and young startups.
apply and so funding outcomes tended to revert to Among the latter, seed firms are typically younger,
their pre-COVID-19 patterns boosted by additional riskier and often more innovative than early-stage
government support in the form of the Recovery Loan and growth firms. The literature suggests seed firms
Scheme. This leads to H1B: usually struggle to secure outside funding, especially
during periods of heightened uncertainty (Baltas
H1B: Initial ECF campaigns launched during the et al., 2022). The substantial challenges encountered
post-COVID-19 period perform similarly to those of by such firms at the onset of the COVID-19 crisis
the pre-COVID-19 period. (January 2020–March 2020) are consistent with these
observations, as noted by Brown et al. (2020). The
The other novel factor that impacted ECF dur- traditional investors in seed firms in normal times are
ing COVID-19 was government intervention on an mostly BA (Block et al., 2019) and VC (Weiss, 2023).
unprecedented scale aimed at reducing capital-raising ECF platforms during the pandemic presented such
barriers for small firms generally but for seed firms investors with two unique advantages for investing
in particular (Calabrese et al., 2022; Cowling et al., directly in seed firms.
2023; Feyen et al., 2020; OECD, 2020). The UK First and crucially, investors could immediately
government supported SMEs financially through acquire direct equity stakes in seed firms via ECF
schemes like the BBLS, CBILS (Coronavirus Busi- instead of potential deferred equity acquired via con-
ness Interruption Loan Scheme) and others. These vertible notes. While the government support pack-
were aimed at addressing market imperfections by ages exerted an overall positive effect on seed firms,
offering government loan guarantee schemes (LGS) they had one unintended side effect: the resultant debt
to smaller (seed- and early-stage) firms during the increases inflated their leverage ratios, especially at
systemic liquidity crisis resulting from the COVID- a time of falling equity valuations. This made equity
19-induced sudden drop in revenues (Cowling et al., injections even more urgent (Calabrese et al., 2022;
2023). In the UK, more than 1 million businesses Cowling et al., 2023; Jibril et al., 2021). The related
accessed these schemes (Cowling et al., 2023). Cala- second advantage of ECF in this context was that the
brese et al. (2022) find that some 92% of the debt availability of government-backed loans paved the
funding provided during this period was backed by way for a private equity injection by BA of VC. Since
the UK government, swamping the usual support higher-performing micro and small firms were the
level of under 5%. primary beneficiaries of government loans (Calabrese
Jibril et al. (2021) using survey data from the et al., 2022), these firms were able to pass both ECF
SME Finance Monitor for Q3 and Q4 2020 report a platform checks (Kleinert et al., 2022) and lead inves-
positive short-term effect of the policy instruments tor due diligence in raising initial ECF funds during
employed by the UK government. Wilson et al. the COVID-19 period. Fulghieri et al. (2020) offer
(2023) observed that patterns of insolvency during another rationale that applies in this context. Firms
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1685
that already have debt are more likely to use equity in Walthoff-Borm et al., 2018). Several control variables
follow-up offerings. This leads to H2B on the outper- are included to account for the observable heteroge-
formance of seed firms. neity in firms seeking to raise ECF funds. Ln(Goal
(£m)) refers to the natural logarithm of the goal (tar-
H2B: Seed firms exhibit improved ECF per- get) set at the beginning of the campaign. Under the
formance compared to other ECF firms during All-or-Nothing provision point mechanism, startups
COVID-19. can retain the raised capital if and only if they reach
or exceed this threshold (Cong & Xiao, 2024). Equity
offered (%) is a reliable signal of the entrepreneur’s
3 Methodology confidence in the firm. Ceteris paribus, the more con-
fident she is about her firm (project), the smaller the
The COVID-19 pandemic was an exogenous equity share she is prepared to sell to outsiders (Ahl-
shock that provides a singular research opportu- ers et al., 2015; Vismara, 2016).
nity for examining the capital-raising performance Firm-related characteristics include Pre-money
of firms on ECF platforms. This study employs a Valuation (£m), which is the agreed valuation of a
categorical variable regression to analyse differ- firm (jointly by the founder(s) and platform) prior to
ences in performance between campaigns in the the initial ECF campaign and is a proxy for the size
pre-COVID-19, COVID-19 and post-COVID-19 (Coakley et al., 2022a). Firm size is positively associ-
periods. Variations in ECF performance are ated with the ability to survive disasters or crises (Bal-
given by the logged values of the Amount raised, tas et al., 2022). Firm stage is a categorical variable
number of Funders and Overfunding (Amount/ for three types of young firms involved in ECF cam-
Goal) outcomes. All regressions take account of paigns: seed, early stage and growth stage. Founder
industry-fixed effects. The categorical variable Team size can affect ECF outcomes and the literature
(COVIDcat) takes the value of 1 for campaigns suggests that a higher number of founders is associated
conducted during the pre-COVID-19 period (Janu- with higher success in ECF (Coakley et al., 2022b).
ary 2018–February 2020), 2 for campaigns dur- An Enterprise Investment Scheme (Weiss) tax relief
ing COVID-19 (March 2020–December 2021) dummy, which equals one for firms using this scheme
and 3 for post-COVID-19 campaigns (December and zero otherwise (Vu & Christian, 2023), captures
2021–October 2023). A dummy variable (LGS_d), potential professional investor involvement. The num-
taking the value of 1 for the duration of the gov- ber of Views and Followers of the campaign on the
ernment’s COVID-19 loan schemes in 2020–2021, Crowdcube platform is used to proxy for the social
is employed to examine their effects.5 capital and visibility of firms. This has an important
The empirical setting of this study comprises 660 signalling role from the investor’s perspective (Vis-
initial ECF offerings on the Crowdcube platform from mara, 2016) and a visibility role from the firm’s view-
January 2018 to October 2023. Campaigns that offer point (Walthoff-Borm et al., 2018). Finally, industry
convertibles or debt (bonds) are not included. The dummies are used to control for industry-fixed effects
dataset comprises just initial ECF campaigns as fol- and capture the unobservable traits of each industry.
low-on ECF campaigns can potentially benefit from
a positive certification effect from a successful initial 3.1 Regression models
campaign (Coakley et al., 2022a). The choice of the
platform and the sampling criteria are in line with The estimation method employed is ordinary least
most previous ECF studies (e.g. Cerpentier et al., squares (OLS) and robust standard errors are
2022; Coakley et al., 2022a, 2024; Estrin et al., 2018; reported in all tables. The following model is used
to test hypotheses H1A and H1B, where the ECF
performance for each campaign is proxied by the
5
As part of a robustness check, additional analysis incorpo-
rating the interaction of LGS_d and COVIDcat was conducted.
The results are consistent with those presented in our main
model in Table 6.
Vol.: (0123456789)
1686 S. Kazembalaghi et al.
natural logarithm of Amount, Funders and Over- using a COVID-19 dummy to examine the results of
funding (Amount/Goal) ratio6: hypotheses H1A and H1B. Here, the comparison is
( ) between the COVID-19 sample (Covid_d = 1) period
Ln ECF performancei = 𝛼1 + 𝛽1 COVIDcat campaigns and the remaining sample campaigns
+ Γ1 Controlsi + 𝜀1 (Covid_d = 0). As a robustness test of H2A, Model (2)
(1) is re-estimated for the first year of COVID-19 (March
where i denotes campaign i and Controlsi represents a 2020–December 2021), contrasting the period when
vector of control variables as defined in Table 8 in the government loan schemes were available with all other
Appendix. The categorical variable COVIDcat can be periods. Propensity score matching (PSM) is employed
rewritten in the form of two dummies with associated to examine the robustness of the H2B results. This
𝛽11 and 𝛽12 coefficients, respectively: method has been used by several ECF and COVID-19
studies to deal with endogeneity (Coakley et al., 2022b;
Li et al., 2023; Vismara, 2019). The use of PSM here
( )
Ln ECF performancei = 𝛼1 + 𝛽11 COVID_Pre
+ 𝛽12 Post_Pre + Γ1 Controlsi + 𝜀1 seeks to answer the following question: are seed firms
(1.1) during the COVID-19 period, ceteris paribus, less/
The coefficients 𝛽11 and 𝛽12 respectively capture more likely to outperform when compared with coun-
the initial and net impacts of COVID-19, compared to terfactuals in the non-COVID-19 period matched on
the performance metrics of the pre-COVID-19 base Goal (£m), Equity (%), Pre-money Valuation (£m) and
period. Team Size.7 Here the sample includes seed firms.
The following two models are employed to test
hypotheses H2A and H2B:
4 Data and empirical analysis
Ln (ECF performancei ) = 𝛼2 + 𝛽2 LGS_d
(2) The section analyses 660 initial ECF campaigns on
+ Γ2 Controlsi + 𝜀2
Crowdcube, the leading UK-based ECF platform.
( )
Ln ECF performancei = 𝛼3 + 𝛽31 Seed_d + 𝛽32 COVID_d
Data were collected from the platform using a cus-
tomised program designed for information scraping.
+ 𝛽33 Seed_d × Covid_d + Γ3 Controlsi + 𝜀3
Because the process is automated, it is largely free of
(3)
potential human error. When the necessary informa-
The primary explanatory variable in Model (2) tion is unavailable, the dataset is supplemented by
is LGS_d which is equal to 1 for the period during data obtained from Companies House (Cerpentier
which the UK government loan schemes (BBLS and et al., 2022; Vismara, 2019). The dataset covers three
CBILS) were offered and 0 otherwise (both during distinct periods, as shown in Table 1.
the COVID-19 period). The main coefficient of inter- This gives the distribution of campaigns across three
est in Model (3) is 𝛽33 which is the interaction of seed (approximately) 2-year sample periods. Nearly one-
firm campaigns with the COVID-19 period. This third (33.66%) of the campaigns were conducted dur-
shows the effect of being a seed stage firm for ECF ing the actual COVID-19 period, with a greater num-
initial campaign outcomes during COVID-19 com- ber occurring pre-COVID-19 and fewer afterwards. It
pared to the non-COVID-19 period. also shows that government LGS were available for
over half (54%) of COVID-19 period campaigns.
3.2 Robustness analysis
4.1 Descriptive statistics
Several robustness tests were performed to evaluate
the robustness of the results. Model (1) is re-estimated Table 2 reports the summary statistics for the vari-
ables defined in Table 8 in the Appendix.
6
The estimation results of this model using a post-COVID-19
dummy (post-COVID-19 campaigns vs. COVID-19 cam-
7
paigns) are presented in Table 13 in the Appendix for the The post-estimation results of this method are available in
robustness tests of H1A and H1B. Table 14 and Figure2 in the Appendix.
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1687
Table 1 Campaign No. campaigns Percent
distribution. Sub-sample
periods Pre-COVID-19 January 2018–February 2020 256 40.65%
COVID-19 March 2020–December 2021 209 33.66%
LGS available 113
LGS unavailable 96
The COVID-19 period Post-COVID-19 December 2021–October 2023 195 25.69%
ended in mid-December Total 660
2021
Table 2 Descriptive statistics Table 2 highlights the presence of right-skewness
for the outcome variables Amount, Funders and
Variable Mean Median
Overfunding, as their means are all substantially
Amount (£m) 0.79 0.48 greater than their corresponding medians. Hence,
Funders (k) 0.67 0.38 these variables are logged in the empirical analysis
Overfunding 1.82 1.43 to mitigate the impact of outliers. Table 2 shows that
Goal (£m) 0.44 0.3 firms offer, on average, 11.11% (median of 9.37%)
Equity (%) 11.11 9.37 of their equity capital, implying that founders clearly
Pre-money Valuation (£m) 12.25 4.53 retain majority control. The median Pre-money Valu-
Team size 1.35 1 ation is £4.53 m, and the median Team Size is 1, indi-
Firm stage 1.95 2 cating that solo founders predominate. Approximately
EIS 0.73 1 73% of the campaigns include an EIS option to attract
Followers (k) 1.21 0.7 professional investors through tax relief schemes.
Views (k) 28.49 19.26 Figure 1 shows the mean value of the ECF cam-
This table displays the summary statistics for the variables paign outcomes and the campaign goals over the
used in our empirical analysis three COVID-19 periods.
Fig. 1 ECF outcomes (pre-, COVID-19 and post-COVID-19). the pre-, COVID-19 and post-COVID-19 periods. The dataset
Note: This bar chart depicts the mean and median values for includes successful initial ECF campaigns on Crowdcube from
Amount (£m), Funders (k), Overfunding and Goal (£m) for January 2018 to May 2023
Vol.: (0123456789)
1688 S. Kazembalaghi et al.
Table 3 Firm stage and COVID-19
Pre-COVID-19 campaigns COVID-19 campaigns Post-COVID-19 campaigns All campaigns
(N) (%) (N) (%) (N) (%) (N) (%)
Seed 48 26% 81 39% 20 14% 149 28%
Early-stage 96 52% 81 39% 92 63% 269 50%
Growth-stage 42 23% 45 22% 33 23% 120 22%
Total 186 100% 207 100% 145 100% 538 100%
This table presents the number (%) of campaigns based on firm stage for the pre-COVID-19, COVID-19 and post-COVID-19 periods
The total COVID-19 campaigns is 207 as firm stage data are missing for 2 firms
The Amount raised in initial campaigns increased Table 4 displays the test results for equality of
in the pre-COVID-19 period and continued to grow mean and median (nonparametric Pearson chi-
thereafter. Simultaneously, the average funding goal square) for COVID-19 vs. pre-COVID-19 and post-
(the amount requested by ventures) is on the rise. COVID-19 vs. pre-COVID-19 variables. Panel A
This ongoing increase in funding goals chimes with shows that the COVID-19 mean and median values of
the Beauhurst (2022) observation that the ECF mar- Funders, Overfunding and Pre-money Valuation are
ket is increasingly attracting more established firms. significantly larger than their pre-COVID-19 counter-
As the market evolves, it attracts an increased pres- parts at the 5% level or better (the mean Amount is
ence of professional investors (Coakley et al., 2024) also considerably larger). Panel B shows that both the
alongside the crowd. The observed decrease in the post-COVID-19 mean and median values of Amount,
median value of funding goal during the COVID-19 Goal and Pre-money Valuation are significantly larger
period, contrasted with an increase in its mean value, than their pre-COVID-19 levels at a 5% significance
can be attributed to a higher number of successful level or better, while the Equity (%) share sold is
initial campaigns by seed firms during the pandemic, significantly smaller. The mean and median differ-
while (large) outliers have concurrently driven up the ence of Funders and degree of Overfunding are both
average value. The COVID-19 period attracted the insignificant.
highest number of Funders and the greatest degree
of Overfunding (Amount/Goal) and both exhibit con- 4.2 Regression results
cave patterns over the sample period.
Table 3 provides a breakdown of ventures run- Table 5 presents the results of the categorical vari-
ning initial ECF campaigns into seed, early-stage able regression models. These serve to evaluate
and growth categories. Prior to COVID-19, more the performance of ECF campaigns over both the
than half (52%) of the ventures were early stage, COVID-19 and post-COVID-19 periods relative
with the remaining firms distributed between seed to the baseline pre-COVID-19 period. The pri-
(26%) and growth-stage (23%) categories. The mary explanatory variables are COVID_Pre and
proportion of seed firms jumped to 39% during Post_Pre in these models. The former gives the
the pandemic but fell sharply to just 14% during COVID-19 effect or the difference between the
the post-pandemic period. Correspondingly, the COVID-19 and pre-COVID-19 periods. The coef-
share of early-stage firms rose from 39% during ficient on Post_Pre gives the difference between the
COVID-19 to some 63% post-COVID-19. The post-COVID-19 and pre-COVID-19 periods, or the
share of growth-stage firms remained stable at net COVID-19 effect. The dependent variables are
22–23% over the sample period. Table 9 in the the performance metrics Ln(Amount), Ln(Funders)
Appendix presents the pairwise correlation coeffi- and Ln(Overfunding) in Models 1 and 2, 3 and 4,
cients between all variables included in the regres- and 5 and 6, respectively. The second model in each
sion analysis. Except for the first three columns pair (for example, Model 1 in the Model 1 and 2
involving the dependent variables, none of the pair) adds industry-fixed effects (FE) to the relevant
pairwise correlations exceed 0.6. regression.
Vol:. (1234567890)
Table 4 Equality of means and median tests
Panel A. COVID-19 vs. pre-COVID-19
COVID-19 Pre-COVID-19
Variables Mean Median Mean Median Mean difference Median difference
Amount (£m) 0.82 0.48 0.66 0.41 0.16** 0.07
Funders (k) 0.79 0.47 0.55 0.34 0.24*** 0.13***
Overfunding 2.07 1.58 1.72 1.42 0.35*** 0.16***
Goal (£m) 0.42 0.25 0.38 0.3 0.04 -0.05
Equity (%) 12.79 9.24 11.83 11.78 0.96 -2.54**
Pre-money Valuation 11.89 4.31 7.58 3 4.31** 1.31**
(£m)
Team size 1.4 1 1.31 1 0.09 0
Firm Stage 1.83 2 1.97 2 -0.14* 0
Digital equity and government support during COVID‑19
EIS 0.68 1 0.72 1 -0.04 0
Followers (k) 1.38 0.84 1.12 0.65 0.25 0.19*
Views (k) 32.5 19.33 28.75 20.87 3.76 -1.54
Panel B. (Post-COVID-19 vs. pre-COVID-19)
Post-COVID-19 Pre-COVID-19
Variables Mean Median Mean Median Mean difference Median difference
Amount (£m) 0.93 0.567 0.66 0.413 0.27*** 0.154***
Funders (k) 0.7 0.371 0.55 0.335 0.15 0.036
Overfunding 1.66 1.303 1.72 1.424 − 0.06 − 0.121
Goal (£m) 0.54 0.4 0.38 0.3 0.16*** 0.1***
Equity (%) 8.27 6.31 11.83 11.78 − 3.56*** − 5.47***
Pre-money Valuation (£m) 18.97 7.5 7.58 3 11.39*** 4.5***
Team size 1.35 1 1.31 1 0.04 0
Firm stage 2.09 2 1.97 2 0.12* 0
EIS 0.78 1 0.72 1 0.06 0
Followers (k) 1.05 0.6 1.12 0.65 − 0.11 − 0.05
Views (k) 22.61 16.15 28.75 20.87 − 5.96* − 4.72***
Significance at the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively
Vol.: (0123456789)
1689
1690 S. Kazembalaghi et al.
Table 5 ECF campaigns: COVID-19 vs. pre-COVID-19, post- vs. pre-COVID-19 results
(1) (2) (3) (4) (5) (6)
Variables Ln(Amount (£m)) Ln(Amount (£m)) Ln(Funders (k)) Ln(Funders (k)) Ln(Overfunding) Ln(Overfunding)
COVID_Pre 0.139*** 0.136*** 0.273*** 0.281*** 0.117*** 0.122***
(3.24) (3.12) (4.89) (5.05) (2.62) (2.66)
Post_Pre 0.0661 0.0734 − 0.0063 − 0.00428 − 0.00927 0.00229
(1.41) (1.60) (− 0.0857) (− 0.0612) (− 0.208) (0.05)
Ln(Goal (£m)) 0.840*** 0.835*** 0.222*** 0.192***
(29.10) (29.05) (5.16) (4.96)
Equity (%) 0.00350** 0.00357** − 0.000147 − 0.00103 0.00148 0.00156
(2.49) (2.29) (− 0.0519) (− 0.355) (1.14) (1.10)
Pre-money Valuation (£m) 0.00049 0.000628 − 0.000887 − 0.000927 − 0.000605 − 0.000416
(0.39) (0.50) (− 0.486) (− 0.539) (− 0.634) (− 0.429)
Team size − 0.00927 − 0.0058 0.00107 0.00702 − 0.0156 − 0.00964
(− 0.402) (− 0.249) (0.03) (0.22) (− 0.651) (− 0.392)
Firm stage 0.0539* 0.0566* 0.0626 0.0639 − 0.021 − 0.02
(1.87) (1.91) (1.48) (1.64) (− 0.814) (− 0.758)
EIS 0.0970** 0.0924** 0.125** 0.091 0.0821** 0.0752*
(2.54) (2.39) (2.24) (1.63) (2.13) (1.93)
Views (k) 0.000587 0.000667 0.00554** 0.00570*** 0.000209 0.000219
(0.67) (0.77) (2.40) (2.62) (0.28) (0.30)
Followers (k) 0.111*** 0.103*** 0.213*** 0.196*** 0.107*** 0.100***
(4.07) (4.05) (3.43) (3.56) (4.74) (4.68)
Constant − 0.122 − 0.0349 5.592*** 5.528*** 0.319*** 0.459***
(− 1.089) (− 0.209) (33.77) (28.82) (4.41) (2.90)
Industry FE No Yes No Yes No Yes
Observations 487 487 487 487 487 487
Adjusted R-squared 0.851 0.852 0.596 0.642 0.222 0.228
The dependent variable in Models (1), (3) and (5) is the natural logarithms of Amount, Funders and Overfunding (Amount/Goal),
respectively, and the key coefficients of interest are those of COVID_Pre and Post_Pre. The estimation method is ordinary least
squares (OLS). Models (2), (4) and (6) are the same as (1), (3) and (5), respectively, but include industry-fixed effects. The standard
errors are robust. The dataset consists of successful initial ECF campaigns on Crowdcube from January 2018 to October 2023. Sig-
nificance at the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively
Virtually, all small firm capital gearing ratios were Amount raised was a considerable £790 k (£480 k)
sharply inflated during COVID-19 by a combination for the full sample.
of increasing debt levels and lower equity valuations The COVID-19 effect (COVID_Pre) is signifi-
as a result of increasing risk levels and little or no cantly positive at the 1% level in Models (1) and (2),
retained earnings. Debt levels rose due to the liquid- indicating that the COVID-19 period campaigns
ity crisis from falling sales revenue and the need for raised a significantly larger Amount (of equity) than
additional funds, including from government-backed their pre-COVID-19 counterparts. In Model (2) with
loans. Note that the Amount of ECF funds raised industry FE, the coefficient is 0.136, suggesting that
directly increases a firm’s equity and, ceteris paribus, Amount raised by COVID-19 campaigns is 14.6% (e.136
improves its gearing ratio. Thus, the Amount raised is – 1 = 0.146) higher than their pre-COVID-19 counter-
the most salient performance metric in our regression parts. Such considerable levels of new equity helped
analysis. It should be noted that the mean (median) ventures in reducing gearing ratios. This supports H1A.
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1691
The performance variable used in Models (3) from 2022Q2 (Kacer & Wilson, 2023).8 Summing
and (4) is Ln(Funders). The COVID-19 effect up, the Table 5 regression results show that initial
(COVID_Pre) is significant at the 1% level in ECF campaigns exhibited an unexpected surge in
Model (4). Its value implies that the number of funding performance during the full COVID-19
funders during COVID-19 was some 32% (e.281 period relative to the pre-COVID-19 period which
– 1 = 0.324) higher than that in the pre-COVID-19 returned to pre-COVID-19 levels in the post-
period. This is an impressive increase in funders COVID-19 period.
and likely reflects large numbers of crowd inves- The coefficients on Ln(Goal) are significantly
tors attracted by the increasing involvement of BA positive at the 1% level in Models (1) to (4), indi-
and VC investors in ECF campaigns. Models (5) cating that Goal exerts a positive impact on both the
and (6) employ Ln(Overfunding) as the depend- Amount raised and the number of Funders. Ln(Goal)
ent variable. The COVID-19 effect (COVID_Pre) is excluded from Models (5) and (6) due to endoge-
in Eq. (6) is significantly positive at the 1% level. neity concerns. The other two control variables that
This implies that COVID-19 campaigns enjoyed are mostly significant across models are EIS and Fol-
a 13% (e.122 – 1 = 0.13) increase in the Overfund- lowers. The significantly positive coefficients of EIS
ing ratio relative to pre-COVID-19 campaigns. in Models (1) to (6), but excluding (4), underline the
Both Model (4) and (6) results also clearly support direct impact of (tax relief for) professional investors
H1A. Our findings for the full pandemic period on Amount raised, Funders and Overfunding. The
extend the results of Cumming et al. (2021b) and control variable, Followers, is significantly positive at
Vu and Christian (2023) for the early months of the 1% level in all regressions underlining the impor-
COVID-19 in the USA and UK, respectively, to tant impact of campaign visibility (Walthoff-Borm
the full COVID-19 period. They showcase the et al., 2018) and social capital (Vismara, 2019; Vu &
outperformance of ECF campaigns during the full Christian, 2023) on performance.
COVID-19 period using three different metrics. Overall, the results reported in Table 5 reveal
The increases reported above range from 13% for a heightened interest in equity capital on ECF plat-
Overfunding and 14.6% for Amount raised to 32% forms, suggesting a surge in private equity demand
for Funders. during the crisis. This uptick coincides with the
Finally, note that the coefficients on Post_Pre lockdown-induced constraints on traditional entre-
are statistically insignificant across all six regres- preneurial finance sources like BA and VC (Mason &
sions in Table 5. These imply that the post- Botelho, 2021), and the scarcity of funding for small
COVID-19 funding performance was not signifi- firms (Baltas et al., 2022; Cumming et al., 2021b;
cantly different from the pre-COVID-19 funding Zhang et al., 2015). They also underline the positive
performance, indicating a return to the status quo impact of COVID-19 on ventures’ use of ECF as a
ante. This supports H1B. These results provide viable source of equity finance (Cumming & Rear-
initial evidence of the post-pandemic resilience don, 2022; Vu & Christian, 2023).
of the ECF ecosystem to the COVID-19 shock Table 6 compares ECF campaign outcomes dur-
and the agility of digital platforms in adapting to ing the April 2020–March 2021 period, when both
increased demands for outside equity from smaller the BBLS and CBILS government LGS were avail-
firms. The fact that ECF performance mean able, versus the second year of COVID-19 when they
reverted to pre-Covid levels rather than worsen were withdrawn. An LGS dummy variable (LGS_d)
was unexpected. We conjecture that government is employed to capture this effect. The coefficient of
support in the form of the Recovery Loan Scheme LGS_d is positive and highly significant across all
introduced in March 2021 may have helped some regressions, indirectly suggesting a positive effect of
ECF firms to remain at normal (Pre-COVID-19) the BBLS (and CBILS) government loan guarantee
levels of business. Note that post-COVID-19 ECF schemes, which backed most of SME debt finance
performance contrasts with that of professional
equity investment that enjoyed a boost during
COVID-19, peaked in 2022Q1 before declining 8
We are grateful to an anonymous reviewer for this point.
Vol.: (0123456789)
1692 S. Kazembalaghi et al.
Table 6 Loan guarantee scheme (LGS) vs. non LGS campaigns
(1) (2) (3) (4) (5) (6)
Variables Ln(Amount (£m)) Ln(Amount (£m)) Ln(Funders (k)) Ln(Funders (k)) Ln(Overfunding) Ln(Overfunding)
LGS_d 0.0916 0.133** 0.206** 0.232*** 0.151*** 0.184***
(1.62) (2.29) (2.52) (2.86) (2.67) (3.10)
Ln(Goal (£m)) 0.796*** 0.784*** 0.219*** 0.184***
(16.75) (14.73) (3.46) (2.84)
Equity (%) 0.00423** 0.00390* 0.00105 − 0.00029 0.000898 0.0008
(2.32) (1.71) (0.41) (− 0.0837) (0.54) (0.36)
Pre-money Valuation (£m) 0.0036 0.0036 0.00380* 0.00314 0.00154 0.00163
(1.53) (1.38) (1.66) (1.31) (0.78) (0.76)
Team size − 0.00073 − 0.0082 0.0449 0.022 − 0.00651 − 0.00129
(− 0.0254) (− 0.293) (0.95) (0.48) (− 0.201) (− 0.0391)
Firm stage 0.0422 0.0473 0.00346 0.0171 − 0.0715* − 0.0631
(0.85) (0.91) (0.06) (0.31) (− 1.838) (− 1.580)
EIS 0.219*** 0.212*** 0.182** 0.164* 0.174*** 0.145**
(3.59) (3.18) (2.16) (1.87) (2.88) (2.25)
Views (k) − 0.000426 − 0.000208 0.00262 0.00322* − 0.00052 − 0.000538
(− 0.597) (− 0.268) (1.55) (1.86) (− 0.804) (− 0.753)
Followers (k) 0.126** 0.114** 0.226** 0.198** 0.122*** 0.115**
(2.56) (2.35) (2.42) (2.32) (2.81) (2.52)
Constant − 0.196 − 0.249 5.764*** 5.827*** 0.357*** 0.51
(− 1.233) (− 0.667) (30.09) (16.87) (3.78) (1.61)
Industry FE No Yes No Yes No Yes
Observations 207 207 207 207 207 207
Adjusted R-squared 0.85 0.852 0.617 0.652 0.257 0.265
The dependent variables in Models (1), (3) and (5) are the natural logarithms of Amount, Funders and Overfunding (Amount/Goal),
respectively, where the key coefficient of interest is that of a Loan Guarantee Scheme dummy, LGS_d. The estimation method is
ordinary least squares (OLS). Models (2), (4) and (6) represent the same models as (1), (3) and (5), respectively, but they include
industry-fixed effects. The standard errors are robust. The dataset includes successful initial ECF campaigns for the full COVID-19
period (March 2020–Dec 2021). Significance at the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively
during the first year of COVID-19 (Calabrese et al., to examine the performance of seed versus more
2022). The coefficient of 0.133 in Model (2) indi- established firms during COVID-19.9 The coeffi-
cates that, during the government support period, cients of Covid_d align with our previous regres-
campaigns raised 14.2% (e0.133 – 1) higher Amounts sion results. The negative Seed_d coefficient is
of funds than during the second year of COVID- significant at the 5% level in (6), insignificant in
19. Model (4) suggests that during the first year of (4) and significant at the 10% level only in (2).
COVID-19, the number of investors was 26.1% These indicate that seed firms generally underper-
(e0.232 – 1) higher than that in the second year. Along form compared to more established firms which
similar lines, the Model (6) Overfunding coefficient is in line with the typical information asymme-
implies a 20.2% (e0.184 – 1) increase in the Amount- try problems of smaller firms (Coakley & Lazos,
to-goal ratio in the first relative to the second year of 2021; Wilson et al., 2018). In Model (2) with
COVID-19. These results highlight the overall posi-
tive impact of the UK government’s support of small
firms during COVID-19 (Calabrese et al., 2022;
9
Jibril et al., 2021) and support H2A. The robustness of the analysis is confirmed by examining
Table 7 reports the results of the interaction of both seed firms exclusively and in combination with early-
stage firms to mitigate any influence the platform’s definition
a COVID-19 period dummy with a seed dummy of ‘seed firms’ might have on our findings.
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1693
Table 7 COVID-19 period and seed firm ECF outcomes
(1) (2) (3) (4) (5) (6)
Variables Ln(Amount (£m)) Ln(Amount (£m)) Ln(Funders (k)) Ln(Funders (k)) Ln(Overfunding) Ln(Overfunding)
Covid_d 0.190* 0.0645 0.376*** 0.229** 0.105** 0.0757
(1.90) (1.46) (3.83) (2.82) (2.15) (1.61)
Seed_d − 0.696*** − 0.141* − 0.394*** − 0.0923 − 0.114** − 0.159**
(− 6.381) (− 1.849) (− 3.890) (− 0.708) (− 2.450) (− 1.984)
Covid_d × Seed_d − 0.144 0.139* 0.00624 0.179 0.162** 0.168**
(− 0.873) (1.85) (0.04) (1.41) (1.99) (2.10)
Ln (Goal (£m)) 0.847*** 0.195***
(31.31) (4.82)
Equity (%) 0.00354** − 0.00113 0.00157
(2.25) (− 0.680) (1.08)
Pre-money Valuation (£m) 0.000819 − 0.00109 − 0.000393
(0.66) (− 0.929) (− 0.406)
Team size − 0.00489 0.00721 − 0.0098
(− 0.208) (0.18) (− 0.390)
Firm stage 0.016 0.0635 − 0.06
(0.33) (0.84) (− 1.270)
EIS tax 0.0866** 0.0832 0.0653*
(2.19) (1.41) (1.66)
Views (k) 0.000626 0.00581*** 0.000341
(0.77) (4.29) (0.47)
Followers (k) 0.102*** 0.197*** 0.101***
(4.13) (5.86) (4.68)
Constant − 0.551*** 0.155 6.020*** 5.590*** 0.475*** 0.611***
(− 9.830) (0.81) (113.10) (20.63) (18.78) (3.14)
Industry FE No Yes No Yes No Yes
Observations 538 487 538 487 538 487
Adjusted R-squared 0.122 0.853 0.064 0.643 0.03 0.236
This table presents the OLS regression results with robust standard errors of the interaction of Seed_d and Covid_d. Covid_d ×
Seed_d is the main variable of interest. The dependent variables in models (1), (3) and (5) are the natural logarithms of Amount,
Funders and Overfunding (Amount/Goal), respectively. Models (2), (4) and (6) add control variables and industry-fixed effects to
previous models. The dataset includes successful initial ECF campaigns on Crowdcube from January 2018 to October 2023. Signifi-
cance at the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively
Ln(Amount (£m)) as the dependent variable, the The findings in Tables 5, 6 and 7 are consist-
interaction term coefficient is positive but signifi- ent with BA and VC funds leveraging the due dili-
cant at the 10% level only, while it is insignificant gence of ECF platforms and of the lead investor
in (4). However, it is statistically significant at the in campaigns to invest in seed startups. ECF plat-
5% level in Model (6), suggesting that seed firms forms enable them to gain a direct equity stake in
enjoy an 18.2% (e0.168 – 1) increase in Overfund- seed firms rather than an indirect one via convert-
ing relative to early-stage and growth firms dur- ible loan notes that they traditionally employed.
ing COVID-19. Overall, COVID-19 had a positive The findings are also consistent with an element of
effect on Overfunding, and, at a lower significance the ‘spray and pray’ approach of VC funds diver-
level, the Amount raised by seed firms in compari- sifying their approach to seed firms with smaller
son to their more established counterparts. These investments and limited involvement in govern-
results lend some support to H2B. ance (compared to their traditional approach)
Vol.: (0123456789)
1694 S. Kazembalaghi et al.
across more startups over the past decade or so that, during COVID-19, seed firms outperformed
in the USA (Ewens et al., 2018). Moreover, these their counterparts in non-COVID-periods. These
funds were attracted by digital ECF platform gov- findings are consistent with those in Table 7, dem-
ernance and, where relevant, the quality signal onstrating that during the COVID-19 period, seed
of LGS loans to invest in seed firms. They are in firms’ access to ECF was unrestricted and they sig-
line with the findings of Jibril et al. (2021) on the nificantly outperformed.
positive impact of the policy instruments used by
the UK government to support SMEs (Calabrese
et al., 2022; Cowling et al., 2023; Wilson et al., 6 Conclusions
2023). Here, the prospect of using ECF campaigns
for debt and capital gearing ratio reduction likely This paper provides clear evidence of an unexpected
led to additional successful campaigns during surge in ECF performance during the COVID-19
COVID-19. Indeed, this observation aligns with period relative to that of the pre-COVID-19 period.
the finding that ECF platforms emerged as the One explanation for this is the ECF platform evolu-
most active investors in early-stage equity in the tion towards attracting high-quality firms, a task at
UK in 2020 (Beauhurst, 2022; Cerpentier et al., which it was less successful during its early years
2022). (Blaseg et al., 2021; Walthoff-Borm et al., 2018).
The pandemic’s constraints inadvertently tipped the
scales in favour of digital equity platforms, eclipsing
5 Robustness tests traditional avenues such as BA and VC. This pivot is
compounded by a broader, seismic shift from public
This section discusses the results of robustness to private equity, reshaping the investment landscape
tests. First, Table 10 in the Appendix presents over recent decades.
the robustness test results for H1A and H1B. The These unexpected findings in this context are in
Covid_d coefficient compares the performance of step with the burgeoning and positive impact of digi-
COVID-19 period campaigns versus non- (pre- talization on entrepreneurial finance which has cre-
and post-) COVID-19 campaigns. These results ated new financial avenues that complement tradi-
reconfirm that ventures attract significantly larger tional intermediaries as discussed by Vismara (2022).
Amounts of capital, more Funders and higher Over- Digitalisation serves a dual purpose. First, it empow-
funding ratios during the COVID-19 than the non- ers nascent firms, even those with limited financial
COVID-19 periods. track records and no audited accounts, to secure
Table 11 in the Appendix presents the results of financial resources through internet platforms (Butt-
comparing the first year of COVID-19 with all other icè & Vismara, 2022; Inceoglu et al., 2024). This cor-
periods. The coefficient of the government LGS_d is roborates the democratizing influence of ECF (Cum-
positive and significant in the models, reconfirming ming et al., 2021a; Fisch et al., 2022) and its leading
the H2A hypothesis. role during the COVID-19 period (Cumming et al.,
Table 12 in the Appendix presents the results of 2021b). The latter period offered a unique labora-
utilising propensity score matching to compare the tory setting for testing the performance of ECF at a
performance of seed firms during the COVID-19 time of heightened uncertainty and limited access for
period with those in the other periods. For robust- smaller firms to traditional entrepreneurial finance
ness purposes, this analysis is conducted using two sources due to lockdown restrictions (Baltas et al.,
routines in Stata—psmatch2 (Panel A) and teffects 2022; Cumming et al., 2021b; Zhang et al., 2015).
(Panel B)—which utilize different approaches for Second, the disruption brought by the pandemic,
calculating standard errors. The Average Treat- unlike that of natural disasters, precipitated a shift
ment Effect on Treated (ATET), capturing the in due diligence practices as digital funding became
causal impact of COVID-19 on seed performance, a necessity (Cumming et al., 2021a). This situation,
is reported using one, three and five matches per marked by intensified information asymmetries,
observation. The significantly positive coefficients favoured digital channels like ECF and marketplace
across all three measures of performance suggest (P2P) and other fintech lending. Consequently, ECF
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1695
platforms thrived by catering to high-calibre small of financial support, they influence firm strategies
private firms that met exacting due diligence stand- in a distinct manner. Savio et al. (2024) discuss how
ards. ECF thus emerged as a lifeline for these ven- debt moratoria—pausing debt payments—can bolster
tures, ensuring their survival and prosperity. For SME long-term growth. Conversely, loan guarantees
traditional and professional investors, the pandemic acted as a short-term buffer against the COVID-19
served as an impromptu nudge for transitioning to crisis by providing much-needed working capital and
online services offered by digital platforms. This liquidity and potentially supporting related ECF fund-
paves the way for future research on the post-pan- ing via a liquidity signal during the pandemic. While
demic online investment tactics of BAs and VCs who Savio et al. (2024) investigate principal-principal con-
have unintentionally test-driven the value added by flicts within a large SME sample, our study suggests
ECF platforms during the COVID-19 pandemic. that such conflicts are less prevalent in digital ECF
This paper reveals an unforeseen positive con- funding, given the equitable share ownership and
sequence of government loan guarantee schemes exit rights ensured by digital nominee shareholding
(LGS), notably the Bounce Back Loan Scheme with (Coakley et al., 2024).
80% repayment guarantees, on equity sources for This study has some limitations. For example,
small firms. Contrary to initial concerns about poten- the dataset is confined to just one of the two major
tial negative effects on SME loans and other finance UK ECF platforms—the Crowdcube platform. How-
sources,10 the findings highlight the beneficial impact ever, this is the longest established and largest UK
on small firms’ access to equity capital—an outcome platform which vouches for the broad applicability
not anticipated at the scheme’s inception. The paper of our findings. Moreover, although more than 90%
underlines the pivotal role of LGS during the initial of SMEs utilized the government guarantee schemes
year of the COVID-19 crisis. By providing typical for raising debt during the first year of the pandemic
repayment guarantees of 80%, these schemes not only (Calabrese et al., 2022), the UK government’s policy
offered vital liquidity support for equity crowdfund- of withholding the identities of state-backed COVID-
ing (ECF) campaigns by recipient firms but signifi- 19 loan recipients11 restricts directly linking our ECF
cantly bolstered the performance of seed firms. This dataset with that for the COVID-19 loans. Nonethe-
outperformance among young seed firms stands out less, the huge numbers of firms that availed of LGS
as particularly surprising, given that traditional inves- schemes and the related increase in their capital
tors typically favour more established firms, a pref- gearing ratios make it plausible that many such firms
erence intensified by the financial crisis (Vismara, sought outside equity from ECF platforms and/or VC
2018). The resilience and adaptability of ECF plat- and BA investors to recapitalize.
forms during COVID-19 further underscore their The findings suggest two areas for future research.
critical role in mitigating the impacts of external dis- First, the empirical results are consistent with ECF as
ruptions. This dual effect of LGS and ECF platforms funding of first resort for small private firms during
in facilitating small firms’ access to equity capital a crisis period. The question is whether this is driven
highlights their importance, offering policymakers by the COVID-19 crisis effect or whether it is a more
evidence of the effectiveness of LGS, particularly in general phenomenon for young, high-quality private
supporting seed firms. firms with growth aspirations. Second, it would be
This paper provides a complementary perspective interesting to explore the relative performance dur-
to that of Savio et al. (2024) who analyse government ing the COVID-19 period of ventures that received a
financial support mechanisms in Italy from 2008 to combination of ECF, and other digital funding (e.g.
2014, specifically examining temporary debt suspen- market invoice finance) compared to ventures that
sion. Our study focuses on loan guarantees. While received ECF funding only. However, this currently
both debt moratoria and loan guarantees are forms would face data challenges.
10 11
Please refer to https://assets.publishing.service.gov.uk/ https://www.theguardian.com/business/2023/jan/05/names-
media/5ee11ef3d3bf7f1eb4a1b4eb/200501_AO_Direction_let- of-uk-covid-business-loan-applicants-to-stay-secret-tribunal-
ter_on_Bounce_Back_Loans_Scheme.pdf rules
Vol.: (0123456789)
1696 S. Kazembalaghi et al.
Appendix
Tables 8, 9, 10, 11, 12, 13 and 14.
Figure 2
Table 8 List of variables
Dependent variables
Amount (£m) Total Amount raised in the campaign
Funders (k) The number of funders (investors) at the end of the campaign
Overfunding Amount raised divided by goal
Explanatory variables
COVIDcat A categorical variable that takes 1 for campaigns pre-COVID-19, 2 during COVID-19 and 3 post-
COVID-19
COVID_Pre A dummy variable that takes 1 for the COVID-19 period and zero for pre-COVID-19
Post_Pre A dummy variable that takes 1 for the post-COVID-19 period and zero for pre-COVID-19
Covid_d A dummy variable that takes 1 for the COVID-19 period and zero otherwise
LGS _d A dummy variable that takes 1 for the period BBLS/CBILS was offered (April 2020–March 2021) by
the UK government and 0 otherwise
Seed_d A dummy variable that takes 1 for seed firms and zero for early-stage and growth firms
Other variables
Ln(Goal (£m)) The natural logarithm of funding Goal that firms set at the beginning of a campaign
Equity (%) Equity (%) of firm’s equity issued during the campaign
Pre-money Valuation (£m) Firm valuation (£m) before the crowdfunding campaign
Team size The number of founders
Firm stage A categorical variable that takes 1 for seed-, 2 for early- and 3 for growth-stage
EIS A dummy that takes 1 if firms use the Enterprise Investment Scheme tax relief and zero otherwise
Views (k) Number of viewers of the firm on the Crowdcube platform
Followers (k) Number of followers of the firm on the Crowdcube platform
Vol:. (1234567890)
Table 9 Correlation matrix
Variables
Amount (£m) 1
Funders (k) 0.780* 1
Overfunding 0.467* 0.511* 1
COVIDcat 0.123* 0.063 − 0.012 1
LGS_d − 0.015 0.046 0.182* 0.04 1
Digital equity and government support during COVID‑19
Goal (£m) 0.754* 0.372* − 0.026 0.153* − 0.097* 1
Equity (%) 0.03 − 0.052 0.048 − 0.118* 0.144* 0.061 1
Pre-money Valuation (£m) 0.594* 0.655* 0.215* 0.176* − 0.05 0.436* − 0.243* 1
Team size 0.024 0.059 0.046 0.028 0.007 0.003 0.136* − 0.014 1
Firm stage 0.429* 0.298* 0.041 0.057 − 0.104* 0.444* − 0.073* 0.380* − 0.077* 1
EIS − 0.126* − 0.084* − 0.003 0.054 − 0.055 − 0.106* − 0.068 − 0.199* 0.07 − 0.073* 1
Followers (k) 0.700* 0.933* 0.461* − 0.025 0.053 0.315* − 0.012 0.570* 0.082* 0.216* − 0.074 1
Views (k) 0.540* 0.686* 0.245* − 0.064 0.106* 0.374* 0.077* 0.532* 0 0.302* − 0.157* 0.586* 1
This table presents the correlation matrix detailing the relationships among all variables utilized in this study. Significance at the 10%, 5% and 1% levels is indicated by *, ** and
***, respectively
Vol.: (0123456789)
1697
1698 S. Kazembalaghi et al.
Table 10 COVID-19 vs. non-COVID-19 period campaigns (Robustness H1A and H1B)
(1) (2) (3) (4) (5) (6)
Variables Ln(Amount (£m)) Ln(Amount (£m)) Ln(Funders (k)) Ln(Funders (k)) Ln(Overfund) Ln(Overfund)
Covid_d 0.107*** 0.101*** 0.276*** 0.283*** 0.121*** 0.120***
(3.10) (2.80) (5.35) (5.79) (3.34) (3.28)
Ln (Goal (£m)) 0.848*** 0.844*** 0.221*** 0.191***
(31.37) (31.18) (5.44) (5.23)
Equity (%) 0.00349** 0.00356** − 0.000147 − 0.00103 0.00147 0.00161
(2.51) (2.30) (− 0.0517) (− 0.355) (1.14) (1.13)
Pre-money Valuation (£m) 0.000658 0.00081 − 0.000903 − 0.000937 − 0.000638 − 0.000356
(0.54) (0.66) (− 0.501) (− 0.555) (− 0.681) (− 0.376)
Founding team size − 0.00826 − 0.00455 0.000974 0.00694 − 0.0158 − 0.00801
(− 0.361) (− 0.197) (0.03) (0.22) (− 0.658) (− 0.340)
Firm stage 0.0537* 0.0563* 0.0626 0.0639 − 0.0215 − 0.0185
(1.86) (1.90) (1.48) (1.64) (− 0.838) (− 0.717)
EIS tax relief 0.0995*** 0.0957** 0.125** 0.0908 0.0816** 0.0789**
(2.59) (2.46) (2.25) (1.64) (2.11) (2.13)
Number of views (k) 0.00046 0.000526 0.00555** 0.00570*** 0.000225 0.00035
(0.55) (0.65) (2.42) (2.66) (0.30) (0.48)
Number of followers (k) 0.110*** 0.102*** 0.213*** 0.196*** 0.107*** 0.0899***
(4.15) (4.15) (3.45) (3.58) (4.75) (4.15)
Constant − 0.0786 0.00734 5.588*** 5.526*** 0.316*** 0.458***
(− 0.794) (− 0.046) (− 36.41) (− 29.98) (− 4.628) ( − 2.924)
Industry FE No Yes No Yes No Yes
Observations 487 487 487 487 487 487
Adjusted R-squared 0.85 0.852 0.597 0.643 0.224 0.224
The dependent variables in models (1), (3) and (5) are the natural logarithms of Amount(£m), Funders and Overfunding (Amount/
Goal), respectively, with the key coefficient of interest being Covid_d. The estimation method used is ordinary least squares (OLS).
Models (2), (4) and (6) are the same as (1), (3) and (5), respectively, but include industry-fixed effects. The standard errors are
robust. The dataset includes successful initial campaigns on Crowdcube from January 2018 to October 2023. Significance at the
10%, 5% and 1% levels is indicated by *, ** and ***, respectively
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1699
Table 11 Loan guarantee scheme (LGS) vs. non LGS campaigns (Robustness H2A)
(1) (2) (3) (4) (5) (6)
Variables Ln(Amount (£m)) Ln(Amount (£m)) Ln(Funders (k)) Ln(Funders (k)) Ln(Overfunding) Ln(Overfunding)
LGS_d 0.134*** 0.127*** 0.310*** 0.309*** 0.170*** 0.165***
(2.94) (2.67) (5.66) (5.51) (3.62) (3.35)
Ln (Goal (£m)) 0.855*** 0.849*** 0.235*** 0.201***
(31.46) (31.33) (5.61) (5.35)
Equity (%) 0.00315** 0.00327** − 0.000864 − 0.00165 0.00112 0.00124
(2.22) (2.06) (− 0.306) (− 0.568) (0.85) (0.85)
Pre-money Valuation (£m) 0.000717 0.000884 − 0.000785 − 0.000775 − 0.000475 − 0.000253
(0.58) (0.71) (− 0.444) (− 0.470) (− 0.502) (− 0.259)
Team size − 0.00522 − 0.00214 0.00874 0.0134 − 0.0119 − 0.00641
(− 0.228) (− 0.0938) (0.24) (0.42) (− 0.496) (− 0.263)
Firm stage 0.047 0.0501* 0.045 0.0465 − 0.0245 − 0.0239
(1.64) (1.70) (1.09) (1.20) (− 0.958) (− 0.919)
EIS 0.0998*** 0.0970** 0.125** 0.0937* 0.0835** 0.0780**
(2.63) (2.52) (2.26) (1.70) (2.19) (2.02)
Views (k) 0.000376 0.000454 0.00539** 0.00558*** 0.000109 0.00012
(0.45) (0.57) (2.36) (2.63) (0.15) (0.17)
Followers (k) 0.111*** 0.102*** 0.216*** 0.199*** 0.108*** 0.101***
(4.22) (4.23) (3.52) (3.67) (4.86) (4.83)
Constant − 0.0424 0.0175 5.683*** 5.558*** 0.332*** 0.451***
(− 0.431) (0.11) (36.91) (30.28) (4.90) (2.87)
Industry FE No Yes No Yes No Yes
Observations 487 487 487 487 487 487
Adjusted R-squared 0.851 0.852 0.594 0.639 0.231 0.236
The dependent variables in models (1), (3) and (5) are the natural logarithms of Amount (£m), Funders (k) and Overfunding
(Amount/goal), respectively, with the key coefficient of interest being LGS_d. The estimation method used for these three models
was ordinary least squares (OLS). Models (2), (4) and (6) represent the same models as (1), (3) and (5), respectively, but with the
inclusion of industry-fixed effects. The standard errors are robust. The dataset includes successful initial ECF campaigns on Crowd-
cube from January 2018 to October 2023. Significance at the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively
Vol.: (0123456789)
1700 S. Kazembalaghi et al.
Table 12 Propensity Score
Matching (Robustness Panel A: Results using psmatch2 routine in Stata
H2B) (1) (2) (3)
Ln(Amount) Ln(Funders) Ln(Overfunding)
ATET (One match per observation) 0.04 0.27* 0.17*
(0.24) (1.66) (1.86)
ATET (Three matches per observation) 0.09 0.39*** 0.194***
(0.61) (3.02) (2.59)
ATET (Five matches per observation) 0.15 0.38*** 0.209***
This table reports the
Average Treatment Effect (1.09) (3.01) (2.92)
on Treated (ATET) and the Obs 149 149 149
t-statistics in parentheses Panel B: Results using teffects routine in Stata
using two routines in
Ln(Amount) Ln(Funders) Ln(Overfunding)
Stata. Seed firms during
COVID-19 are matched ATET (One match per observation) 0.348*** 0.647** 0.243*
with one, three and five (7.27) (2.55) (1.88)
counterfactuals based on ATET (Three matches per observation) 0.389** 0.525*** 0.227**
their propensity scores.
(2.46) (3.62) (2.26)
The treatment and control
groups are matched based ATET (Five matches per observation) 0.423*** 0.535*** 0.232***
on the Goal (£m), Equity (2.91) (4.54) (2.66)
(%), Pre-money Valuation Obs 143 143 143
(£m) and Team Size
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1701
Table 13 ECF post-COVID-19 vs. COVID-19 outcomes
(1) (2) (3) (4) (5) (6)
Variables Ln(Amount (£m)) Ln(Amount (£m)) Ln(Funders (k)) Ln(Funders (k)) Ln(Overfunding) Ln(Overfunding)
Post_COVID − 0.0771* − 0.0842* − 0.281*** − 0.314*** − 0.126*** − 0.134***
(− 1.861) (− 1.946) (− 3.842) (− 4.747) (− 3.055) (− 3.109)
Ln (Goal (£m)) 0.839*** 0.842*** 0.215*** 0.210***
(23.44) (21.68) (3.89) (4.17)
Equity (%) 0.00308** 0.00290* 0.000341 − 0.000832 0.00114 0.00104
(2.33) (1.92) (0.12) (− 0.277) (0.91) (0.72)
Pre-money Valuation (£m) 0.000476 0.000761 0.0000371 0.000212 − 0.000459 − 0.0000795
(0.36) (0.56) (0.02) (0.12) (− 0.430) (− 0.0709)
Team size − 0.0146 − 0.0123 − 0.000994 0.00408 − 0.0219 − 0.0151
(− 0.616) (− 0.514) (− 0.0248) (0.11) (− 0.850) (− 0.570)
Firm stage 0.0601 0.0527 0.0394 0.0239 − 0.0272 − 0.0323
(1.59) (1.32) (0.70) (0.48) (− 0.859) (− 0.999)
EIS tax relief 0.154*** 0.149*** 0.143* 0.106 0.132*** 0.121***
(3.44) (3.16) (1.96) (1.45) (2.95) (2.61)
Views (k) 0.000716 0.00075 0.00528** 0.00528** 0.000433 0.000419
(0.76) (0.86) (2.14) (2.38) (0.51) (0.54)
Followers (k) 0.108*** 0.0994*** 0.199*** 0.180*** 0.107*** 0.0985***
(3.72) (3.81) (3.14) (3.34) (4.09) (4.15)
Constant − 0.0221 0.0724 5.900*** 6.143*** 0.416*** 0.564***
(− 0.171) (0.33) (31.02) (23.00) (5.22) (2.88)
Industry FE No Yes No Yes No Yes
Observations 352 352 352 352 352 352
Adjusted R-squared 0.841 0.844 0.56 0.622 0.266 0.283
The dependent variables in Models (1), (3) and (5), are the natural logarithm of Amount (£m), Funders (k) and Overfunding, respec-
tively, with the key coefficient of interest being that on Post_COVID. This is a dummy that takes 1 for the post-COVID-19 period
and zero for the COVID-19 period. The estimation method is ordinary least squares (OLS). Models (2), (4) and (6) are the same as
models (1), (3) and (5), respectively, but add industry fixed effects. The standard errors are robust. The dataset includes successful
initial ECF campaigns for COVID-19 (March 2020–Dec 2021) and post-COVID-19 (Dec 2021–Oct 2023) periods. Significance at
the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively
Vol.: (0123456789)
1702 S. Kazembalaghi et al.
Table 14 PSM post-estimation tests
Variable Unmatched Mean treated Mean control %bias %bias reduction t P >| t| V(T)/ V(C)
matched
Goal (m) U 0.22179 0.2735 − 21.3 − 1.3 0.196 0.75
M 0.19889 0.20563 − 2.8 87 − 0.22 0.83 0.75
Equity (%) U 13.322 12 .212 11.8 0.7 0.484 3.56*
M 12.305 12 .54 − 2.1 82.1 − 0.19 0.847 1.14
Pre-money Valuation (£m) U 3.7744 3.5233 4.3 0.26 0.798 3.81*
M 2.9955 2.8172 3.1 29 0.4 0.688 0.87
Team size U 1.4321 1.3824 6 0.36 0.717 1.41
M 1.4487 1.4197 3.5 41.6 0.22 0.827 1.5
Sample Ps R2 LR χ2 p > χ2 Mean bias Median bias B R %Var
Unmatched 0.031 6.42 0.17 1.9 8.9 39.5* 1.78 50
Matched 0.4 0.77 0.942 2.9 2.9 14 1.16 0
This table reports the t-statistics and p-value for the difference in mean between the treated and control groups before and after
matching. The bias percentage is reported to examine covariate imbalance before and after matching
Fig. 2 Propensity score of
treated and untreated firms
before and after match-
ing. The figure on the left
shows the propensity score
of Treated (seed firms
during COVID-19) versus
Untreated (seed firms in
the non-COVID-19 period)
before matching, whereas
the figure on the right
shows it after matching.
The propensity score refers
to the probability of a firm
being in the Treated (seed
firms during COVID-19)
or Untreated (seed firms in
the non-COVID-19 period)
group given the covariates
calculated based on Goal
(£m), Equity (%), Pre-
money Valuation (£m) and
Team Size
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1703
Acknowledgements An earlier version of this paper was entrepreneurs? Entrepreneurship Theory and Practice,
presented at the 2nd International Fintech Conference in Car- 45, 505–530. https://doi.org/10.1177/10422587198994
diff in November 2023. We are grateful to the participants for Block, J. H., Colombo, M. G., Cumming, D. J., & Vismara,
their helpful comments. We also wish to thank two anonymous S. (2018). New players in entrepreneurial finance and
reviewers whose insightful comments and helpful suggestions why they are there. Small Business Economics, 50, 239–
helped substantially to improve the exposition and overall 250. https://doi.org/10.1007/s11187-016-9826-6
quality of this paper. Finally, we thank the Editor Prof. Paul Block, J., Fisch, C., Vismara, S., & Andres, R. (2019). Pri-
Momtaz for giving us the opportunity to revise our paper. vate equity investment criteria: An experimental con-
joint analysis of venture capital, business angels, and
Funding Open access funding provided by Università degli family offices. Journal of Corporate Finance, 58, 329–
studi di Bergamo within the CRUI-CARE Agreement. 352. https://doi.org/10.1016/j.jcorpfin.2019.05.009
Block, J. H., Groh, A., Hornuf, L., Vanacker, T., & Vismara,
S. (2021). The entrepreneurial finance markets of the
future: A comparison of crowdfunding and initial coin
Open Access This article is licensed under a Creative Com-
offerings. Small Business Economics, 57, 865–882.
mons Attribution 4.0 International License, which permits
https://doi.org/10.1007/s11187-020-00330-2
use, sharing, adaptation, distribution and reproduction in any
Brown, R., & Cowling, M. (2021). The geographical impact
medium or format, as long as you give appropriate credit to the
of the Covid-19 crisis on precautionary savings, firm
original author(s) and the source, provide a link to the Crea-
survival and jobs: Evidence from the United Kingdom’s
tive Commons licence, and indicate if changes were made. The
100 largest towns and cities. International Small Busi-
images or other third party material in this article are included
ness Journal, 39, 319–329. https://doi.org/10.1177/
in the article’s Creative Commons licence, unless indicated
0266242621989326
otherwise in a credit line to the material. If material is not
Brown, R., & Rocha, A. (2020). Entrepreneurial uncer-
included in the article’s Creative Commons licence and your
tainty during the Covid-19 crisis: Mapping the temporal
intended use is not permitted by statutory regulation or exceeds
dynamics of entrepreneurial finance. Journal of Busi-
the permitted use, you will need to obtain permission directly
ness Venturing Insights, 14, e00174. https://doi.org/10.
from the copyright holder. To view a copy of this licence, visit
1016/j.jbvi.2020.e00174
http://creativecommons.org/licenses/by/4.0/.
Brown, R., Rocha, A., & Cowling, M. (2020). Financing
entrepreneurship in times of crisis: Exploring the impact
of COVID-19 on the market for entrepreneurial finance
References in the United Kingdom. International Small Business
Journal, 38, 380–390. https://doi.org/10.1177/02662
42620937464
Ahlers, G. K., Cumming, D., Günther, C., & Schweizer, D.
Butticè, V., & Vismara, S. (2022). Inclusive digital finance:
(2015). Signaling in equity crowdfunding. Entrepreneur-
The industry of equity crowdfunding. Journal of Tech-
ship Theory and Practice, 39, 955–980. https://doi.org/
nology Transfer, 47, 1224–1241. https://doi.org/10.
10.1111/etap.12157
1007/s10961-021-09875-0
Altig, D., Baker, S., Barrero, J. M., Bloom, N., Bunn, P., Chen,
Calabrese, R., Cowling, M., & Liu, W. (2022). Understand-
S., Davis, S. J., Leather, J., Meyer, B., & Mihaylov, E.
ing the dynamics of UK Covid-19 SME financing. Brit-
(2020). Economic uncertainty before and during the
ish Journal of Management, 33, 657–677. https://doi.
COVID-19 pandemic. Journal of Public Economics, 191,
org/10.1111/1467-8551.12576
104274. https://doi.org/10.1016/j.jpubeco.2020.104274
Cerpentier, M., Vanacker, T., Paeleman, I., & Bringmann,
Baker, S. R., Bloom, N., Davis, S. J., & Terry, S. J. (2020).
K. (2022). Equity crowdfunding, market timing,
Covid-induced economic uncertainty. National Bureau of
and firm capital structure. The Journal of Technol-
Economic Research.
ogy Transfer, 47, 1766–1793. https://doi.org/10.1007/
Bali, T. G., Cakici, N., & Whitelaw, R. F. (2011). Maxing
s10961-021-09893-y
out: Stocks as lotteries and the cross-section of expected
Chandler, J. A., Short, J. C., & Wolfe, M. T. (2021). Finding
returns. Journal of Financial Economics, 99, 427–446.
the crowd after exogenous shocks: Exploring the future of
https://doi.org/10.1016/j.jfineco.2010.08.014
crowdfunding. Journal of Business Venturing Insights, 15,
Baltas, K., Fiordelisi, F., & Mare, D. S. (2022). Alternative finance
e00245. https://doi.org/10.1016/j.jbvi.2021.e00245
after natural disasters. British Journal of Management, 33,
Coakley, J., & Lazos, A. (2021). New developments in equity
117–137. https://doi.org/10.1111/1467-8551.12516
crowdfunding: A review. Review of Corporate Finance, 1,
Beauhurst. (2022). The state of UK equity crowdfunding in
341–405. https://doi.org/10.1561/114.00000008
2022. Available at: https://www.beauhurst.com/blog/uk-
Coakley, J., Lazos, A., & Liñares-Zegarra, J. M. (2022a). Sea-
equitycrowdfunding/
soned equity crowdfunded offerings. Journal of Corpo-
Belitski, M., C. Guenther, A. S. Kritikos and R. Thurik (2022).
rate Finance, 77, 101880. https://doi.org/10.1016/j.jcorp
’Economic effects of the COVID-19 pandemic on entrepre-
fin.2020.101880
neurship and small businesses’, Small Business Economics,
Coakley, J., Lazos, A., & Liñares-Zegarra, J. M. (2022b).
pp. 1–17 https://doi.org/10.1007/s11187-021-00544-y
Equity crowdfunding founder teams: Campaign success
Blaseg, D., Cumming, D., & Koetter, M. (2021).
and venture failure. British Journal of Management, 33,
Equity crowdfunding: High-quality or low-quality
286–305. https://doi.org/10.1111/1467-8551.12494
Vol.: (0123456789)
1704 S. Kazembalaghi et al.
Coakley, J., D. J. Cumming, A. Lazos and S. Vismara (2024). Fisch, C., Meoli, M., & Vismara, S. (2022). Does block-
’Syndicated equity crowdfunding and the collective action chain technology democratize entrepreneurial finance?
problem’, Available at SSRN 4703001, https://doi.org/10. An empirical comparison of ICOs, venture capital, and
2139/ssrn.4703001 REITs. Economics of Innovation and New Technology,
Cong, L. W., & Xiao, Y. (2024). Information cascades and 31(1–2), 70–89. https://doi.org/10.1080/10438599.2020.
threshold implementation: Theory and an application to 1843991
crowdfunding. Journal of Finance, 79(1), 579–629. Fulghieri, P., Garcia, D., & Hackbarth, D. (2020). Asymmet-
Cortés, K. R., & Strahan, P. E. (2017). Tracing out capital ric information and the pecking (dis) order. Review of
flows: How financially integrated banks respond to natural Finance, 24, 961–996. https://doi.org/10.1093/rof/r faa0
disasters. Journal of Financial Economics, 125, 182–199. 05
https://doi.org/10.1016/j.jfineco.2017.04.011 Guenther, C., Johan, S., & Schweizer, D. (2018). Is the
Cowling, M., Brown, R., & Lee, N. (2021). The geography of crowd sensitive to distance?—How investment decisions
business angel investments in the UK: Does local bias differ by investor type. Small Business Economics, 50,
(still) matter? Environment and Planning a: Economy and 289–305. https://doi.org/10.1007/s11187-016-9834-6
Space, 53(5), 1180–1200. Inceoglu, I., Vanacker, T., & Vismara, S. (2024). Digitalization
Cowling, M., Wilson, N., Nightingale, P., & Kacer, M. (2022). and resource mobilization. British Journal of Manage-
Predicting future default on the Covid-19 bounce back ment. https://doi.org/10.1111/1467-8551.12817
loan scheme: The£ 46.5 billion question. International Jibril, H., Roper, S., & Hart, M. (2021). Covid-19, business
Small Business Journal, 40, 650–666. https://doi.org/10. support and SME productivity in the UK. ERC Research
1177/02662426211058050 Paper (p. 94). Available at: https://www.productivity.
Cowling, M., P. Nightingale and N. Wilson (2023). ’COVID-19 ac.uk/wp-content/uploads/2021/06/COVID19-business-
lending support and regional levelling up: Evidence from support-and-SME-productivity-inthe-UK-FINAL.pdf
UK loan guarantee schemes’, Regional Studies, pp. 1–16 Kacer, M. and N. Wilson (2023). ’Supporting innovative start-
https://doi.org/10.1080/00343404.2022.2159021 up and growing businesses: Equity finance provision
Cumming, D., Meoli, M., & Vismara, S. (2021a). Does through the pandemic: Interim report’, Available at SSRN
equity crowdfunding democratize entrepreneurial 4456252, https://doi.org/10.2139/ssrn.4456252
finance? Small Business Economics, 56, 533–552. Kleinert, S., Volkmann, C., & Grünhagen, M. (2020). Third-
https://doi.org/10.1007/s11187-019-00188-z party signals in equity crowdfunding: The role of prior
Cumming, D. and R. S. Reardon (2022). ’COVID-19 and financing. Small Business Economics, 54, 341–365.
entrepreneurial processes in US equity crowdfund- https://doi.org/10.1007/s11187-018-0125-2
ing’, Journal of Small Business Management, pp. 1–24 Kleinert, S., Bafera, J., Urbig, D., & Volkmann, C. K. (2022).
https://doi.org/10.1080/00472778.2022.2051178 Access denied: How equity crowdfunding platforms use
Cumming, D. J., A. Martinez-Salgueiro, R. S. Reardon and quality signals to select new ventures. Entrepreneurship
A. Sewaid (2021b). ’COVID-19 bust, policy response, Theory and Practice, 46, 1626–1657. https://doi.org/10.
and rebound: Equity crowdfunding and P2P versus 1177/10422587211011945
banks’, The Journal of Technology Transfer, pp. 1–22 Lattanzio, G., W. L. Megginson and A. Sanati (2023). ’Dis-
https://doi.org/10.1007/s10961-021-09899-6 secting the listing gap: Mergers, private equity, or regu-
Estrin, S., Gozman, D., & Khavul, S. (2018). The evolu- lation?’, Journal of Financial Markets, p. 100836 https://
tion and adoption of equity crowdfunding: Entrepre- doi.org/10.1016/j.finmar.2023.100836
neur and investor entry into a new market. Small Busi- Li, T., Trinh, V. Q., & Elnahass, M. (2023). Drivers of global
ness Economics, 51, 425–439. https://doi.org/10.1007/ banking stability in times of crisis: The role of corporate
s11187-018-0009-50 social responsibility. British Journal of Management, 34,
Ewens, M., & Farre-Mensa, J. (2022). “Private or pub- 595–622. https://doi.org/10.1111/1467-8551.12631
lic equity? The Evolving Entrepreneurial Finance Mason, C., & Botelho, T. (2021). Business angel investing dur-
Landscape.” Annual Review of Financial Economics, ing the covid-19 economic crisis: Evidence from Scot-
14, 271–293. https://doi.org/10.1146/annurev-finan land. Venture Capital, 23, 321–343. https://doi.org/10.
cial-101821-121115 1080/13691066.2021.2019564
Ewens, M., Nanda, R., & Rhodes-Kropf, M. (2018). Cost of Meoli, M., & Vismara, S. (2021). Information manipulation
experimentation and the evolution of venture capital. in equity crowdfunding markets. Journal of Corporate
Journal of Financial Economics, 128, 422–442. Finance, 67, 101866. https://doi.org/10.1016/j.jcorpfin.
Feyen, E., Gispert, T. A., Kliatskova, T., & Mare, D. S. (2020). 2020.101866
Taking stock of the financial sector policy response Momtaz, P. P. (2023). “The economics of PIPEs”, revisited.
to COVID-19 around the world. World Bank Group, Small Business Economics, 60(1), 59–83.
Finance, Competitiveness, and Innovation Global Prac- Myers, S. C., & Majluf, N. S. (1984). Corporate financing
tice. Available at: https://openknowledge.worldbank.org/ and investment decisions when firms have information
server/api/core/bitstreams/34d67981-ec9f-5310-bd93- that investors do not have. Journal of Financial Eco-
5cabf2ae3966/content nomics, 13, 187–221. https://doi.org/10.1016/0304-
Fisch, C., & Momtaz, P. P. (2020). Institutional investors and 405X(84)90023-0
post-ICO performance: An empirical analysis of inves- Nguyen, L., & Wilson, J. O. (2020). ’How does credit sup-
tor returns in initial coin offerings (ICOs). Journal of ply react to a natural disaster? Evidence from the Indian
Corporate Finance, 64, 101679. Ocean Tsunami’, the European Journal of Finance,
Vol:. (1234567890)
Digital equity and government support during COVID‑19 1705
26, 802–819. https://doi.org/10.1080/1351847X.2018. Vismara, S., Paleari, S., & Ritter, J. R. (2012). Europe’s sec-
1562952 ond markets for small companies. European Financial
OECD. (2020). Coronavirus (COVID-19): SME policy Management, 18(3), 352–388. https://doi.org/10.1111/j.
responses. Available at: https://www.oecd.org/en/publi 1468-036X.2012.00641.x
cations/2020/07/coronavirus-covid-19-sme-policyresp Vu, A. N., & Christian, J. (2023). UK equity crowdfunding
onses_c7ff14a3.html#:~:text=It%20examines%20how% success: The impact of competition, Brexit and Covid-19.
20SMEs%2 0are, respon ses%2 0to%2 0fost er%2 0SME% British Journal of Management. https://doi.org/10.1111/
20resilience 1467-8551.12714
Paterson, A., Sakariyahu, R., Lawal, R., & Alabi, A. (2024). Walthoff-Borm, X., Schwienbacher, A., & Vanacker, T. (2018).
The impact of government policy responses to the COVID- Equity crowdfunding: First resort or last resort? Journal
19 pandemic and Brexit on the UK financial market: A of Business Venturing, 33, 513–533. https://doi.org/10.
behavioural perspective. British Journal of Management, 1016/j.jbusvent.2018.04.001
35, 174–191. https://doi.org/10.1111/1467-8551.12702 Weiss, G. (2023). ’A theory of seed financing’, Available at
Ritter, J., Signori, A., & Vismara, S. (2013). ’Economies SSRN 4668015, https://doi.org/10.2139/ssrn.4668015
of scope and IPO volume in Europe’, In “Handbook of Wilson, N., Wright, M., & Kacer, M. (2018). The equity gap and
Research on IPOs”,pag. 11–34, ed. Levis, M., & Vis- knowledge-based firms. Journal of Corporate Finance, 50,
mara, S. (eds), 2013, Edward Elgar, Cheltenham, UK, 626–649. https://doi.org/10.1016/j.jcorpfin.2017.12.008
ISBN: 978 1 78195 536 9. Wilson, N., M. Kacer and M. Cowling (2023). ’Crisis, loan
Savio, R., Castellaneta, F., Vismara, S., & Zattoni, A. (2024). schemes, bank lending, insolvency and default: An analy-
Exploring the third type of agency problem: An empiri- sis of early outcomes of policy interventions’, Insolvency
cal study of the impact of debt suspension programmes and Default: An Analysis of Early Outcomes of Policy
on SMEs’ resource allocations. British Journal of Man- Interventions.(September 2023). https://doi.org/10.2139/
agement. https://doi.org/10.1111/1467-8551.12795 ssrn.4589327
Vismara, S. (2016). Equity retention and social network the- Wright, M., Hart, M., & Fu, K. (2015). A nation of angels:
ory in equity crowdfunding. Small Business Economics, Assessing the impact of angel investing across the UK.
46, 579–590. https://doi.org/10.1016/j.techfore.2018.07. ERC. Available at: https://www.enterpriseresearch.ac.uk/
014 wp-content/uploads/2019/11/ERC-Angels-Report..pdf
Vismara, S. (2018). Information cascades among investors Zhang, B., Wardrop, R., Rau, P. R., and Gray, M. (2015). Mov-
in equity crowdfunding. Entrepreneurship Theory and ing mainstream: Benchmarking the European alternative
Practice, 42, 467–497. finance market. Journal of Financial Perspectives, 3(3),
Vismara, S. (2019). Sustainability in equity crowdfunding. Tech- 60–77.
nological Forecasting and Social Change, 141, 98–106.
Vismara, S. (2022). Expanding corporate finance perspec- Publisher’s Note Springer Nature remains neutral with regard
tives to equity crowdfunding. Journal of Technol- to jurisdictional claims in published maps and institutional
ogy Transfer, 47, 1629–1639. https://doi.org/10.1007/ affiliations.
s10961-021-09903-z
Vol.: (0123456789)
File and source
- File
- kazembalaghi-2024-digital-equity-and-government-support-during-covid-19.pdf
- Size
- 877,272 bytes
- SHA-256
- 5a9c3f48c6a3c6b1e7723ddf737d4daad905b9d8a2cf2622c57d10e08f815c33
- Original
- link.springer.com